Resumen
So far, business forecasting has been considered important in almost all economic entitiesand it is often used in areas such as in security analysts, institutional lending, and manage-ment. This research aims at examining empirically the predictability of time series of earn-ings for future earnings and stock price patterns by means of Autoregressive Integrated Mov-ing Average (ARIMA). It is expected to provide contribution in the form of empirical evi-dence, in which earnings are considered useful for predicting earnings and stock price pat-tern. The forecasting is by using some techniques among others, the nae model, regression,ARIMA (Box-Jenkins) and so on. The data were taken from stock market data center at UGMand UTYs IDX corner during 1996-2007. Based on the sampling criteria, 22 companieswere used as the sample. The results showed that there were no statistically significant dif-ferences among actual earnings for the earnings forecast. The first hypothesis which statesthat there is ability in predicting earnings income is statistically supported. The second hy-pothesis which states that there is the ability of earnings in predicting stock price pattern isalso statistically supported